6754
J. Am. Chem. SOC.1981, 103, 6754-6755
Scientific Laboratory, jointly supported by the U. S. Department of Energy and the N I H (RR-00962-02, Division of Research Resources). Dr. Chou-Hong Tann and Kazys Martinkus are warmly thanked for expert technical assistance. We are grateful to Dr. Donald Borders, Lederle Laboratories, Pearl River, NY, for generous gifts of streptothricin F and Streptomyces L-1689-23. Peter Demou of the Chemistry Department, Yale University, is thanked for obtaining the I3C NMR spectra. The Bruker HX-270 NMR instrument facility at Yale used in this work was supported by the National Science Foundation, Grant CHE-79 16210, from the Chemistry Division.
A
B
I
(b-H)Fe4(C0)12($-COH): Evidence for a Protonated $-CO Complex as an Intermediate in the Proton-Induced Reduction of CO
3bO
’
I
‘
230
’
Wv+ ’ 260
Figure 1. ”C N M R spectra of (A) ( N - H ) F ~ ~ ( C O ) , ~ ( ~ J ~(I) - C and OH) (B) [PPN][HFe4(CO),,] (11). These spectra were observed a t 20 M H z on a Varian CFT-20 spectrometer at -90 “C in CD2CI2.
K. H. Whitmire and D. F. Shriver* a
Department of Chemistry, Northwestern University Evanston, Illinois 60201 Received June 29. 1981
In their original N M R spectroscopic investigation of metal carbonyls in acid media, Wilkinson and co-workers demonstrated that protonation occurs on the metal.’ Subsequent diffraction studies have confirmed this general concept and provided detailed structural information on the variety of bonding patterns between the proton and metal centers in polynuclear carbonyls.2 In contrast to this earlier work it was recently shown that protonation of some metal carbonyl clusters also may occur at edge-bridging (eq 1)3y4or face-bridging carbonyl oxygen^.^
\I/
p
+
\I/
,OH-C
/
(1)
B
/I\ The present research provides spectroscopic evidence for a new type of 0-protonated carbonyl ligand, q2-COH, resulting from the protonation of [HFe4(CO),,]-. This new 0-protonated compound appears to be a key intermediate in the recently discovered proton-induced reduction of C O in [Fe4(C0)13]2-.6 In 1957, Hieber and Werner reported a compound with the empirical formula H2Fe4(CO)13,which was described as soluble in ethers and benzene and stable for significant periods of time at room temperature? Attempts to structurally characterize this compound in several laboratories have been uniformly unsuccessful,8 so we have explored the possibility that, as with HFe3(CO),,(COH), the anhydrous diprotonated form of the tetranuclear cluster may be stable only at low temperatures. Anhydrous (p-H)Fe4(C0)12(q2-COH)(I) was prepared under an inert atmosphere by the addition of 30 pL of HS03CF3 or H S 0 3 F to ca. 3 mL of a frozen (-196 “C) CD2C12 solution containing 0.13-0.18 mmol of [PPN] [HFe4(CO),,] (enriched to ca. 15% I3CO) in an N M R tube. The tube was sealed under (1) A. Davison, W. McFarlane, L. Pratt, and G. Wilkinson, J. Chem. Soc., 3653 (1962). (2) “Transition Metal Hydrides”, R. Bau, Ed., “Advances in Chemistry Series”, American Chemical Society, Washington, DC, 1978,No. 167. (3) H. A. Hodali, D. F. Shriver, and C. A. Ammlung, J. Am. Chem. SOC., 100, 5239 (1978). (4) J. B. Kiester, J . Orgunomet. Chem., 190, C36 (1980). (5) G.Fachinetti, J . Chem. Soc., Chem. Commun., 397 (1979). (6) K.Whitmire and D. F. Shriver, J . Am. Chem. SOC.,102, 1456 (1980). (7) W. Hieber and R. Werner, Chem. Ber., 90, 286 (1957). (8) B. T.Heaton and P. Chini, private communications.
0002-7863/81/1503-6754$01.25/0
Figure 2. I3C N M R spectra of (A) [PPN][HFe,(CO),,] (11) and (B) (p-H)Fe4(C0)12(~2-COCHJ) (111). Spectra were obtained a t 90 M H z on a Nicolet NT-360 spectrometer a t -90 OC in CD2C12.
vacuum and warmed to -90 OC, and the I3Cand ‘H NMR spectra were determined. Additional I3C N M R spectra were obtained on I3CO enriched samples of [HFe,(CO),,]- (11) and HFe4(C0)12(q2-COCH3)(111), both of which have been the subjects of X-ray structure determination^.^^'^ The q2-C0 in I1 displays a characteristic low-field I3C N M R feature (Figures 1 and 2) at 281 ppm relative to Me4Si. Upon reaction with the methyl carbocation to produce I11 the resonance due to the q2-C0 shifts to even lower field, 301 ppm. Similarly, the protonation of I1 leads to a low-field shift of the resonance of the q2-C0 to 294 ppm, indicating that protonation has occurred (9) M. Manassero, M.Sansoni, and G. Longoni, J . Chem. SOC.,Chem. Commun., 919 (1976). (10)K. Whitmire, D.F. Shriver, and E. M. Holt, J . Chem. SOC.,Chem. Commun., 780 (1980).
0 1981 American Chemical Society
J . Am. Chem. SOC.1981, 103, 6755-6757
on oxygen.11 Further evidence for the existence of the $-COH species is provided by the observation of ‘H NMR resonances of 6 13.2 and -26.5 which are assigned to an oxygen-bound proton and a bridging metal hydride, respectively. The I3C N M R spectrum in the terminal CO region is more complicated, but comparison of the spectra of 1-111 shows consistent trends. The 13C0 resonances of I1 display an intensity pattern of 1:2:1:1:2:2:2:2, consistent with the symmetry found in the crystal structure. That of I11 is similar, 1:2:2:2:2:2:2 (Figure 0
6755
Table I. Proton-Induced Reduction Reactions with and without Agents4 Added amount of complex used, m ol
iron complex [PPN],[Fe,(CO),3] [PPN],[Fe,(CO),] [PPNlz[Fe4(Co)~31 ~[PPN],[Fe,(CO),] [PpNI,[Fe,(CO),,I {[PPN] ,[Fe,(CO),]
1.14 8.14
1.97
x lo-, x 1°-,1
7.65 x lo-’ 1-09 x 6.29 X l O - ’ t
CH,
CO
H,
0.56 trace 0.76
2.2 2.2b 2.2
0.20 0.57b 0.37
1.0
3.0
0.37
a Yields after three days reaction, are given in mol/mol of Fe, cluster. b’ Yield based on moles of Fe,(C0),2-.
a four-iron butterfly with an $-COH was lacking. The present characterization of I adds credence to the original proposal. The conversion of I to another reaction intermediate, HFe4(C0)12($CH) (eq 2), requires 2 equiv of protons and electrons. The / 0
/”
I,X=H 11, X = e111, X = CH,
2). In the case of 111, two of the unique CO’s, b and d, have accidental, near degeneracy. This is supported by the slight splitting in the peak at 212.5 ppm. Compound I shows a pattern of 1:2:1:2:3:4. The resonance of area three is assigned to one of the unique C O S (b or d) which has achieved accidental degeneracy with two other carbonyls. It also is possible that upon protonation a fluctional process causes rapid interchange of these three carbonyls; however, the first hypothesis is more consistent with observations for I1 and 111. The feature of area 4 is assigned to two resonances of area 2, indicated by the asymmetry. These are also unresolved for I1 on the 20-MHz instrument.12 As found for ( p H ) Fe3(CO)lo(F-COH), (p-H)Fe4(CO) 12(42COH) is highly unstable at room temperature in CH2C12solution. Reexamination of the product reported earlier as H2Fe4(C0)137 leads us to question the original formulation. When [Fe4(CO)13]Zwas treated with a large excess of aqueous 6 M HCl according to the method of Hieber and Werner7 and then dried over Na2S04 overnight, the resulting infrared spectrum showed bands characteristic of [HFe4(CO),,]- and H 2 0 or H30+.nH20 (intense broad bands around 3450 and 1630 cm-’). In a different attempted preparation of H2Fe4(C0)13,addition of a small excess of concentrated hydrochloric acid to an ether slurry of [PPN[2[Fe4(CO),,] produced a precipitate of 2 equiv of [PPNICl. The carbonyl stretching region of the spectrum was again identical with that of [HFe,(CO),,]-. Therefore this product is formulated as [H30.nH20][HFe4(CO),,], which is analogous to the previously isolated [H30.nH20][HFe3(C0)11].3When dried under vacuum a brown solid results which display a complex I3CNMR spectrum, indicating the presence of a mixture of iron carbonyls (IR:vco (hexane) 2045 vs, 2020 s, 1998 m, 1985 sh, 1980 sh, 1894 w, 1865 w cm-I). An identical IR spectrum is obtained for “ H 2 F e 4 ( C 0 ) 1 ~ when the synthesis is attempted at room temperature under anhydrous conditions. Thus we conclude that the original “H2Fe4(C0)13”was a complex mixture. Isotope labeling experiments demonstrated that when I11 undergoes proton-induced reduction the carbon monoxide being reduced is derived from the q2-COCH3moiety.1° By analogy it was proposed that I is an intermediate when [Fe4(C0)13]2-is exposed to strong acid,I3 but firm evidence for the existence of (1 1) The shift to lower field in the I3C NMR signal of a bridging carbonyl upon complex formation of the CO oxygen with an electrophile is well documented. See J. R. Wilkinson and L. J. Todd, J . Organomet. Chem., 118, 199 (1976); H. A. Hodali and D. F. Shriver, Inorg. Chem., 18, 1236 (1979). (12) Because of the instability of (p-H)Fe4(C0)12($-COH),its 13CNMR spectrum was obtained at Northwestern University where only a 20-MHz instrument was available. The 90-MHz spectra were obtained at the NSF Regional NMR Facility at the University of Illinois, Urbana, IL. (13) E. M. Holt, K. Whitmire, and D. F. Shriver, J . Organomet. Chem., 213, 125 (1981).
latter appears to be supplied by sacrificial oxidation of an iron SPies, because addition of [Fe2(Co)d2-, which does not Produce CH4 upon reaction with HS03CF3, increases the Yield of CH4 (Table 1)In summary, the present results provide an example of a new type of Protonated carbon “oxide ligand and indicate that the Protonation of a carbon r”xide ligand activates co toward (2-0 bond scission and reduction. Acknowledgment. This research was supported by the National Science Foundation through Grant CHE-79 18010. High-field 13CNMR measurements were made at the University of Illinois NSF Regional Instrumentation Facility, (Grant CHE 97-16100). We appreciate informative correspondence and conversations with Brian Heaton and the late Paolo Chini during our initial attempts to characterize the elusive H2Fe4(CO)13.
Photocatalytic Oxidations of Lactams and N-Acylamines James W. Pavlik* and Supawan Tantayanon Department of Chemistry Worcester Polytechnic Institute Worcester, Massachusetts 01609 Received May 21, 1981
Although there is considerable interest in the use of semiconductors in photovoltaic cells and in the synthesis of high-energy compounds as potential fuels,’ their use as photocatalysts in organic synthesis has received much less consideration.2 As part of our program to explore the utility of semiconductors in synthetic organic photochemistry, we wish to report that the 5- and 6membered lactams and N-acylamines undergo photocatalytic oxidation to the corresponding imides upon irradiation in the presence of oxygenated aqueous suspensions of Ti02. Irradiation of 10 mL of 0.2 M aqueous solutions of amides 1-6 in the presence of 100 mg of suspended unreduced anatase TiOz (1) See, for example: Heller, A,; Miller, B.; Lewerenz, H. J.; Bachmann, K. J. J. Am. Chem. SOC.1980,102,6555. Wrighton, M. S. Acc. Chem. Res. 1979,12,303. Kraeutler, B.; Bard, A. J. J . Am. Chem. SOC.1978,100,5985 and references therein. (2) Kanno, T.; Oguchi, T.; Sakuragi, H.; Tokumaru, K. Tetrahedron Lett. 1980, 467.
0002-7863/81/1503-6755$01.25/0
0 1981 American Chemical Society
